But don't let them use -A (admin notes), -a (approve), -f (force), -l (list),
-U (impersonate another user), or -c or anything else for any foreigh project.

This was preventing edits to previously approved reservations.

Set debugging on for Reserve and ApproveReservation.

We are running another apache server on boss, on port 51369, which
invokes a backend perl script that maps the URL path argument to the
profile, and then calls out to manage_profile to pull from the
repository and update the profile to reflect the new HEAD branch.
Using mod_rewrite in the apache config to restrict URLs to exactly
the one URL that is accepted, modulo the value of the secret token.

I had to refactor a bunch of code in manage_profile to make it easier to
add a new entrypoint for modification from a git repo. This needed to be
done for a long time, I had never cleaned up the original profile
creation code.

On the edit profile web page, there is a new row in the Repository panel
providing the Push URL, and an explanatory help modal.

There is a new slow polling timer that looks for a change to the repo
hash and causes the web page to update in place from the repo, as when a
push hook is invoked and changes the repo.

If an inner boss/ops has to fsck its filesystems, it can take quite
a while and that happens prior to bringing up the network.

push hook URL.

Breakout ports are identified with a ':\d+' suffix in their port
description. This extension of the existing hack for submodules lets
breakout ports be referenced by their ifindex instead of trying to
encode the combination of module, submodule, port, and subport into a
single port number in the database.  This is total crap, and makes me
hate the mod/port DB encoding we have even more.  It is on my TODO list
to move snmpit to using device-specific port identification based on
port description strings instead.

NOTE: The module number semantics have changed for the submod/subport
hack.  Instead of using whatever module the port is on (plus one...),
the module is set to '0'. This is what should be used in the 'wires' DB
table.

on the Mothership.

code to construct a path is duplicated is 5 different places and they
have gone out of sync. We have not allowed users to set the path for
years, but we were still constructing a path in php code and passing it
along. Throw them all away and construct the path the way we want it.

Apply same rules to admins; admins don't know what they are doing
either. I know, I'm an admin.

But if the admin sets the path to start with /usr/testbed/images, throw
that away, and construct proper path in /usr/testbed/images.

for testbed,tbres,utahstud.

This includes the assimilation of clusterd that Leigh did awhile back
as well as a few bug fixes.

up millions of rows and tell us nothing useful.

1. Small bits of cleanup to remove hardwired paths and test for images
   that live in /usr/testbed and need special handling.

2. Make sure that the image directory exists in /proj when taking a
   snapshot of a system image that lives in /usr/testbed since that is
   were we always write new images.

3. When deleting an image, remove the DB state for the highest numbered
   version if it never came ready/released.

I implemented versioning for images make this test the wrong thing to
do. I need to come back and change this, but its not a big deal right
now.

data back to the DB.

Now you can actually run linktest on a topo that involves one or
more persistent blockstore nodes.

Looks like they were mostly being caught in the event system dispatcher
thread before. Let's see if that was causing problems...

1. Fix for deleting profiles from image deletion; failing to find all
   the profiles to delete.

2. More wording changes.

3. List profile versions and links to them.

After noticing that it seems to require around 30 seconds to get
packet counters from all our control net switches (thanks, cisco6!)
my traditional interval of 10 seconds is leading to highly misleading
traffic estimates. So now you can have it average over the actual
interval it took.

Right now I am doing this as an extra operation, it could be rolled
into the initial createsliver.